About

Semantic segmentation, or image segmentation, is the task of clustering parts of an image together which belong to the same object class. It is a form of pixel-level prediction because each pixel in an image is classified according to a category. Some example benchmarks for this task are Cityscapes, PASCAL VOC and ADE20K. Models are usually evaluated with the Mean Intersection-Over-Union (Mean IoU) and Pixel Accuracy metrics.

( Image credit: CSAILVision )

Benchmarks

TREND DATASET BEST METHOD PAPER TITLE PAPER CODE COMPARE

Libraries

Subtasks

Datasets

Latest papers with code

An MRF-UNet Product of Experts for Image Segmentation

12 Apr 2021balbasty/nitorch

While convolutional neural networks (CNNs) trained by back-propagation have seen unprecedented success at semantic segmentation tasks, they are known to struggle on out-of-distribution data.

SEMANTIC SEGMENTATION

11
12 Apr 2021

Predicting Pedestrian Crossing Intention with Feature Fusion and Spatio-Temporal Attention

12 Apr 2021OSU-Haolin/Pedestrian_Crossing_Intention_Prediction

This work addresses the above limitations by introducing a novel neural network architecture to fuse inherently different spatio-temporal features for pedestrian crossing intention prediction.

SEMANTIC SEGMENTATION

3
12 Apr 2021

Visiting the Invisible: Layer-by-Layer Completed Scene Decomposition

12 Apr 2021lyndonzheng/VINV

In this work, we propose a higher-level scene understanding system to tackle both visible and invisible parts of objects and backgrounds in a given scene.

INSTANCE SEGMENTATION SCENE UNDERSTANDING SEMANTIC SEGMENTATION

2
12 Apr 2021

Location-Sensitive Visual Recognition with Cross-IOU Loss

11 Apr 2021Duankaiwen/LSNet

Object detection, instance segmentation, and pose estimation are popular visual recognition tasks which require localizing the object by internal or boundary landmarks.

2D HUMAN POSE ESTIMATION 2D OBJECT DETECTION INSTANCE SEGMENTATION POSE ESTIMATION SEMANTIC SEGMENTATION

6
11 Apr 2021

Bootstrapping Semantic Segmentation with Regional Contrast

9 Apr 2021lorenmt/reco

We present ReCo, a contrastive learning framework designed at a regional level to assist learning in semantic segmentation.

SEMI-SUPERVISED SEMANTIC SEGMENTATION

5
09 Apr 2021

Spatial Feature Calibration and Temporal Fusion for Effective One-stage Video Instance Segmentation

6 Apr 2021MinghanLi/STMask

To further explore temporal correlation among video frames, we aggregate a temporal fusion module to infer instance masks from each frame to its adjacent frames, which helps our framework to handle challenging videos such as motion blur, partial occlusion and unusual object-to-camera poses.

INSTANCE SEGMENTATION SEMANTIC SEGMENTATION VIDEO INSTANCE SEGMENTATION

3
06 Apr 2021

One Thing One Click: A Self-Training Approach for Weakly Supervised 3D Semantic Segmentation

6 Apr 2021liuzhengzhe/One-Thing-One-Click

Point cloud semantic segmentation often requires largescale annotated training data, but clearly, point-wise labels are too tedious to prepare.

3D SEMANTIC SEGMENTATION

2
06 Apr 2021

Weakly supervised segmentation with cross-modality equivariant constraints

6 Apr 2021gaurav104/WSS-CMER

In addition, we add a KL-divergence on the class prediction distributions to facilitate the information exchange between modalities, which, combined with the equivariant regularizers further improves the performance of our model.

DATA AUGMENTATION SEMANTIC SEGMENTATION WEAKLY SUPERVISED SEGMENTATION

1
06 Apr 2021

Hierarchical Image Peeling: A Flexible Scale-space Filtering Framework

4 Apr 2021ForawardStar/HIPe

The importance of hierarchical image organization has been witnessed by a wide spectrum of applications in computer vision and graphics.

SEMANTIC SEGMENTATION

14
04 Apr 2021